Temporal lobe epilepsy (TLE) presents as an incapacitating neurological syndrome comprised of seizures and associated comorbidities. Seizures in up to 35% of patients with TLE are refractory to common forms of treatment, which are likewise largely ineffective to control comorbid conditions. In line with the NINDS Epilepsy Re- search Benchmarks this exploratory research proposal aims to develop an antisense gene therapy vector as a new approach for the targeted treatment of seizures and associated comorbidities in mesial TLE (mTLE). Our proposal is based on the rationale that deficiency of the endogenous anticonvulsant and seizure terminator adenosine (ADO) is a pathological hallmark of the epileptic brain. ADO-levels in the brain are controlled by metabolic clearance through the astrocyte-based enzyme adenosine kinase (ADK), which is pathologically overexpressed in the epileptic brain. Research from our and other laboratories has convincingly shown that therapeutic ADO augmentation or pharmacological ADK inhibition effectively prevent seizures in at least three different rodent models of mTLE, including seizures that are refractory to common antiepileptic drugs. Despite the proven effectiveness of ADO in seizure control, critical gaps for therapy development remain local specificity to avoid side effects of systemic ADO augmentation, and long-term effectiveness. The overarching goal of our application is (i) to develop, optimize, and characterize a final clinical gene therapy candidate to reduce (not suppress) ADK expression in the epileptic brain through antisense technology and thereby to reinstate near to normal ADO function, and (ii) to provide rigorous efficacy data in two clinically relevant rodent models of mTLE. Availability of a final clinical candidate and proof of effectiveness in relevant disease models are pre-requisites to initiate a pre-pre-IND interaction with the FDA, which is the final expected deliverable of this application. Our approach is supported by preliminary data showing effectiveness of ADO augmentation in a mouse model of pharmacoresistant mTLE and the availability of antisense constructs to reduce ADK expression in astrocytes. TLE patients with pharmacoresistant seizures and who are candidates for resection surgery are ideal subjects for a mechanism-based gene therapy aimed at down-regulating ADK within the epileptogenic hippocampus for several reasons: (i) identification of ADK as a rational and effective therapeutic target; (ii) defined focal area of epileptogenicity to e targeted by local gene therapy; (iii) confinement of endogenous ADK expression largely to astrocytes minimizes off-target effects of astrocyte-selective therapy; (iv) adverse ADO effects, though un- expected, could be treated with the ADO receptor antagonist caffeine; (v) reversibility (i.e. surgical resection) of gene therapy, if needed. Our Specific Aims are: (1) Develop a clinical candidate gene therapy product to effectively reduce ADK expression in astrocytes. (2) Perform preclinical efficacy and safety tests in two clinically relevant rodent models of mTLE. (3) Prepare regulatory documents for discussion at a pre-pre-IND meeting with the FDA.